Prognostic significance of immunohistochemical markers in non-small cell lung cancer

Abstract

7211 Background: The purpose of this study is to use a large patient population to identify immunohistochemical (IHC) biomarkers to enable improved prognostication in patients with non-small cell lung cancer (NSCLC). Methods: A tissue microarray was constructed using duplicate 0.6 mm cores of formalin-fixed paraffin embedded tissue blocks from 609 patients with NSCLC. IHC was used to detect 11 biomarkers including EGFR, HER2, HER3, p53, p63, Bcl-1, Bcl-2, TTF-1, CEA, Ch, and SNP. A clinical database was created prospectively at the time of tissue collection. Survival outcomes were obtained from a Provincial Cancer Registry database. Results: Male to female ratio was 400:209; median age 63yrs (range 35–82); median survival 3.5yrs (mean 5.7). All specimens were reviewed: 243 adenocarcinoma (ACA), 272 squamous cell carcinoma (SCC), 35 large cell carcinoma, 32 non-small cell carcinoma NOS, and 6 other (giant cell carcinoma). 21 patients with other histologies were excluded. Survival data for 535 cases was available. As of June 2005, 429 patients (80%) had died; of these 286 (54%) died of lung cancer, 117 (22%) died of other known causes, and for 26 (5%) the cause of death was not available. Bcl-2 (p = 0.007) was the only biomarker to predict better overall survival (OS). Bcl-2 (p = 0.021) and p63 (p = 0.025) were significant for disease specific survival (DSS) in all NSCLC. Analysis of the subgroups indicated that p63 was significant (p = 0.039) for DSS in squamous cell carcinoma (SCC) but not for adenocarcinoma (ACA) (p = 0.81). Bcl-2 was not significant for DSS in either subgroup (p = 0.28 for SCC, p = 0.112 for ACA). EGFR expression was associated with improved DSS in SCC (p = 0.012) but not for ACA. Co-expression of EGFR-HER3 was more likely in SCC then in ACA (p = 0.033). There was no correlation between outcome and any combination or clustering of biomarkers. Conclusions: The biomarkers p63 and Bcl-2 are predictive of DSS in NSCLC. EGFR expression is predictive of DSS in SCC. Sub-classification of NSCLC by histopathology is important as the relevance of some biomarkers (EGFR) would be lost if pooled. p63, Bcl-2, and EGFR may be used as prognostic markers in patients with NSCLC. Co-expression of EGFR-HER3 is more likely in SCC then in ACA. This may help explain the differential response to EGFR inhibitors in SCC versus ACA. No significant financial relationships to disclose.